Bool q — Epoch AI

Bool q — Epoch AI is run by Epoch AI. It has ranked 136 models, of which the catalogue holds 35, scoring from 59 to 89.4 on Score.

measured by Epoch AI · the board itself

Place	Model	Metric	Score
5thof 136	StableBeluga2	Score	89.4
6thof 136	falcon-180B	Score	89
7thof 136	GPT-4o-mini (2024-07-18)	Score	88.7
9thof 136	Llama-2-70b-hf	Score	88.6
19thof 136	Mistral-7B-v0.1	Score	87.4
21stof 136	GPT-3.5 Turbo (older v0613)	Score	87
23rdof 136	Qwen-14B	Score	86.2
26thof 136	gemma-2-9b	Score	85.7
27thof 136	t5-large	Score	85.4
35thof 136	Phi-3.5-MoE-instruct	Score	84.6
43rdof 136	Mistral-7B-Instruct-v0.2	Score	83.2
44thof 136	gemma-7b	Score	83.2
45thof 136	falcon-40b	Score	83.1
48thof 136	Llama 3.1 8B Instruct	Score	82.8
51stof 136	Llama-2-13b	Score	82.4
57thof 136	t5-base	Score	81.4
59thof 136	vicuna-13b-v1.3	Score	80.8
67thof 136	llama-13b	Score	78.7
73rdof 136	Phi-3.5-mini-instruct	Score	78
74thof 136	Llama-2-7b	Score	77.9
83rdof 136	llama-7b	Score	76.5
86thof 136	Mistral 7B	Score	76.4
88thof 136	t5-small	Score	76.4
91stof 136	opt-66b	Score	76
92ndof 136	phi-1_5	Score	75.8
94thof 136	falcon-7b	Score	75.3
98thof 136	xgen-7b-8k-base	Score	74.3
107thof 136	Bloom	Score	70.4
109thof 136	gemma-2b	Score	69.4
117thof 136	gpt-j-6b	Score	65.4
118thof 136	opt-1.3b	Score	65
119thof 136	gpt-neox-20b	Score	64.9
123rdof 136	gpt-neo-2.7B	Score	61.8
124thof 136	gpt2-xl	Score	61.8
131stof 136	stablelm-tuned-alpha-7b	Score	59

Place

Model

Metric

Score

5thof 136

Score

89.4

6thof 136

Score

7thof 136